- Email: [email protected]
Diagnostic value and prognostic implications of early cardiac magnetic resonance in survivors of out-of-hospital cardiac arrest
Accepted Manuscript Diagnostic value and prognostic implications of early cardiac magnetic resonance in survivors of out of hospital cardiac arrest Alessandro Zorzi, MD, PhD, Angela Susana, MD, Manuel De Lazzari, MD, PhD, Federico Migliore, MD, PhD, Giovanni Vescovo, MD, Daniele Scarpa, MD, Anna Baritussio, MD, Giuseppe Tarantini, MD, PhD, Luisa Cacciavillani, MD, PhD, Benedetta Giorgi, MD, Cristina Basso, MD, PhD, Sabino Iliceto, MD, Chiara Bucciarelli Ducci, MD, PhD, Domenico Corrado, MD, PhD, Martina Perazzolo Marra, MD, PhD PII:
S1547-5271(18)30195-4
DOI:
10.1016/j.hrthm.2018.02.033
Reference:
HRTHM 7502
To appear in:
Heart Rhythm
Received Date: 12 December 2017
Please cite this article as: Zorzi A, Susana A, De Lazzari M, Migliore F, Vescovo G, Scarpa D, Baritussio A, Tarantini G, Cacciavillani L, Giorgi B, Basso C, Iliceto S, Bucciarelli Ducci C, Corrado D, Marra MP, Diagnostic value and prognostic implications of early cardiac magnetic resonance in survivors of out of hospital cardiac arrest, Heart Rhythm (2018), doi: 10.1016/j.hrthm.2018.02.033. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Diagnostic value and prognostic implications of early cardiac magnetic
2
resonance in survivors of out of hospital cardiac arrest
3
Alessandro Zorzi, MD, PhD (1); Angela Susana, MD (1); Manuel De Lazzari, MD, PhD (1);
4
Federico Migliore, MD, PhD (1); Giovanni Vescovo, MD (1); Daniele Scarpa, MD (1); Anna
5
Baritussio, MD (1,3); Giuseppe Tarantini, MD, PhD (1); Luisa Cacciavillani, MD, PhD (1);
6
Benedetta Giorgi, MD (2); Cristina Basso, MD, PhD (1); Sabino Iliceto, MD (1); Chiara Bucciarelli
7
Ducci, MD, PhD (3); Domenico Corrado, MD, PhD (1); Martina Perazzolo Marra, MD, PhD (1).
SC
8
RI PT
1
1) Division of cardiology Department of Cardiac, Thoracic and Vascular Sciences, University of Padova, Italy
M AN U
9 10
2) Division of radiology, Department of Medicine, Az. Ospedaliera di Padova, Italy
11
3) Bristol NIHR Cardiovascular Biomedical Research Unit, Bristol Heart Institute, University of Bristol, United Kingdom
12 13
TE D
14 15 16
Correspondence to:
Domenico Corrado, MD, PhD
18
Department of Cardiac, Thoracic and Vascular Sciences
19
University of Padova
20
Via Giustiniani, 2 - 35121 Padova (Italy)
21
Tel. +39 049 8212458; Fax +39 049 8212309
22
e-mail: [email protected]
AC C
EP
17
23 24
Short title: cardiac magnetic resonance findings in survivors of out-of-hospital cardiac arrest
25
Word count (including title page, abstract, text, references, tables and figures legend): 5065
26
Conflicts of interest: none
1
ACCEPTED MANUSCRIPT
ABSTRACT
28
Background: In patients who survived out-of-hospital cardiac arrest (OHCA) it is crucial to
29
establish the underlying cause and its potential reversibility. Objective: We assessed the
30
incremental diagnostic and prognostic role of early cardiac magnetic resonance (CMR) in survivors
31
of OHCA. Methods: Among 139 consecutive OHCA patients, we enrolled 44 (median age 43
32
years; 84% males) patients who underwent coronary angiography and CMR ≤7 days after
33
admission. The CMR protocol included T2-weighted sequences for myocardial edema and late
34
gadolinium-enhancement (LGE) sequences for myocardial fibrosis. Results: Coronary angiography
35
identified obstructive coronary artery disease (CAD) in 18/44 patients, in whom CMR confirmed
36
the diagnosis of ischemic heart disease by demonstrating subendocardial or transmural LGE; the
37
presence of myocardial edema allowed to differentiate between acute myocardial ischemia (N=12)
38
and post-infarction myocardial scar (N=6). Among the remaining 26 patients without obstructive
39
CAD, 19 (73%) showed at CMR dilated cardiomyopathy (N=5), myocarditis (N=4), mitral valve
40
prolapse associated with LGE (N=3), ischemic scar (N=2), idiopathic non-ischemic scar (N=2),
41
arrhythmogenic cardiomyopathy (N=1), hypertrophic cardiomyopathy (N=1) and Tako-Tsubo
42
cardiomyopathy (N=1). In this subgroup, 6/26 (23%) had myocardial edema. During a mean
43
follow-up of 36±17 months, all 18 patients with myocardial edema had an uneventful outcome
44
while 9/26 (35%) without myocardial edema experienced sudden arrhythmic death (N=1),
45
appropriate defibrillator interventions (N=5) and non-arrhythmic death (N=3) (p=0.006).
46
Conclusions: In survivors of OHCA, early CMR with a comprehensive tissue characterization
47
protocol provided additional diagnostic and prognostic value. The identification of myocardial
48
edema was associated with a favorable long-term outcome.
49
Key Words: cardiac arrest; cardiopulmonary resuscitation; cardiac blunt trauma; cardiac magnetic
50
resonance; implantable cardioverter defibrillator; secondary prevention; ventricular arrhythmia.
AC C
EP
TE D
M AN U
SC
RI PT
27
2
ACCEPTED MANUSCRIPT INTRODUCTION
52
Out-of-hospital cardiac arrest (OHCA) due to ventricular fibrillation (VF) or pulseless
53
ventricular tachycardia (VT) can be caused by ischemic and non-ischemic heart diseases. In patients
54
who suffered OHCA, clinical investigations aim at establishing what is the underlying disease and
55
whether the cause of the arrhythmia is reversible. This is particularly relevant for decision to
56
implant a defibrillator (ICD), that is not indicated in patients who experienced VT/VF due to a
57
reversible myocardial injury like that secondary to myocardial ischemia of acute coronary
58
syndromes or myocarditis[1-2]. In this contest, the interpretation of standard investigations such as
59
ECG, echocardiography and coronary angiography may not be straightforward and nonspecific
60
findings secondary to the cardiac arrest and subsequent resuscitation may act as confounders[3-5].
M AN U
SC
RI PT
51
Cardiac magnetic resonance (CMR), in addition to accurate assessment of ventricular
62
function, allows the identification of myocardial tissue changes such as myocardial edema as
63
evidenced by increased signal intensity on T2-weighted sequences and myocardial fibrosis by late
64
gadolinium enhancement (LGE) on post-contrast sequences[6]. In the evaluation of patients who
65
survived OHCA, the combination of sequences for myocardial edema and fibrosis has the potential
66
to distinguish an acute and potentially reversible injury from a chronic and irreversible lesion[7].
67
The aim of the present study was to prospectively evaluate the incremental diagnostic and
AC C
EP
TE D
61
68
prognostic value of early (within one week after admission) CMR, enhanced with an accurate tissue
69
characterization protocol including sequences for myocardial edema and fibrosis, in OHCA
70
survivors.
3
ACCEPTED MANUSCRIPT METHODS
71
72
The study included all survivors of arrhythmic OHCA who were admitted to our institution during the study period 2011-2016 and underwent coronary angiography and CMR within one week
74
of hospitalization. The CMR was performed in all patients who could be weaned from mechanical
75
ventilation, had no neurological impairment and had no contraindications to CMR. The study
76
complies with the Declaration of Helsinki, was approved by the Ethical Committee and all patients
77
agreed to participate.
78
Coronary angiography
SC
RI PT
73
Coronary angiography was performed in all patients: at admission in those with ST-segment
80
elevation at post-resuscitation ECG or in those with high clinical suspicious of ongoing myocardial
81
ischemia; delayed in the others. A culprit coronary lesion was defined as obstructive (≥70%) CAD
82
with TIMI 0/1 flow with abrupt closure, or TIMI 2/3 flow with features suggestive of thrombus or
83
ulcerated plaques, ST segment and T wave changes in the corresponding ECG location, and
84
evidence of matching regional wall motion abnormality on left ventriculogram or
85
echocardiogram[8].
86
Contrast-enhanced cardiac magnetic resonance
TE D
EP
The CMR scan protocol systematically included steady-state free precession sequence (true
AC C
87
M AN U
79
88
FISP) cine images for morpho-functional evaluation, T2-weighted sequences for detection of
89
myocardial edema and post-contrast sequences for detection of LGE. Details are reported in the
90
supplemental file.
91
Follow-up
92
Patients were followed-up with serial outpatient evaluations or with telephonic interviews to
93
determine the alive status and whether they reached the composite arrhythmic end-point including
94
sudden death, cardiac arrest due to VF, sustained VT and appropriate ICD intervention. Sudden 4
ACCEPTED MANUSCRIPT death was defined as any natural death occurring instantaneously or within 60 minutes from
96
symptoms onset. Appropriate ICD intervention was defined as an ICD shock delivered in response
97
to a ventricular tachyarrhythmia >170 bpm and documented by stored intracardiac ECG data. All
98
ICD interventions were adjudicated as appropriate or inappropriate by two electrophysiologists
99
(AZ, DC) based on morphologic features, tachycardia onset, rate stability and atrial electrograms
100
(when available). ICD were programmed according to the MADIT-RIT trial high-rate or delayed-
101
therapy programming[9].
RI PT
95
104
Statistical Analysis
M AN U
103
SC
102
Continuous and categorical variables are expressed as median (with 25th-75th %iles), and n (%), respectively. Because of the small sample size, normality distribution was not assumed.
106
Categorical variables were compared by using the chi-square or Fisher exact test, as appropriate.
107
Continuous data were compared using the Mann-Whitney U test. A p-value of <0.05 was
108
considered statistically significant. Data were analysed with SPSS® version 23 (IBM®).
AC C
EP
TE D
105
5
ACCEPTED MANUSCRIPT
RESULTS
109
Among a consecutive series of 139 patients admitted for OHCA, 37 (27%) died during the
111
acute phase. Among the remaining 102 patients, 58 (42%) were not recruited because of
112
neurological impairment and/or mechanical ventilation (N=52), the presence of a no-MRI
113
compatible pacemaker (N=4) or poor CMR quality images (N=2). Forty-four patients [median age
114
43 years (39-62 years); 84% males] met the enrollment criteria and constituted the definitive study
115
cohort. (Table 1). Coronary angiography was performed at admission in 20 (45%) patients,
116
including 7 (16%) with post-resuscitation ST-segment elevation, and within 4 days in the remaining
117
patients.
M AN U
SC
RI PT
110
118 119
Obstructive CAD
Angiographic evidence of obstructive coronary disease (≥1 coronary stenosis ≥70%) was
121
found in 18 (41%) patients. A clear culprit coronary lesion was identified in 10 patients, 5/5 with
122
single vessel disease and 5/13 with multiple vessels disease.
TE D
120
Cardiac magnetic resonance findings are summarized in Table 2.
124
The presence of myocardial edema at T2-weighted sequences consistent with an acute
125
ischemic injury in 12/18 (67%) patients (Figure 1). In all cases, the regional distribution of
126
myocardial edema was concordant with the territory which was tributary of a coronary artery with
127
obstructive stenosis or occlusion. In 10 patients with an identifiable culprit coronary lesion, the
128
regions of myocardial edema were always concordant with the regional distribution of the culprit
129
coronary artery. Myocardial edema was present in 2 of the 8 patients with multiple vessels disease
130
and no identifiable culprit coronary lesion. The Peak troponin I (normal value <0.017 ug/L) was
131
significantly higher in patients with [12.2 ug/L (2.1-104.8)] compared with those without [2.8 ug/L
132
(1.2-26.1)] LV myocardial edema (p = 0.02).
AC C
EP
123
6
ACCEPTED MANUSCRIPT Left ventricular LGE with a subendocardial or transmural distribution was demonstrated in
134
all patients (Figure 2). Among the 12 patients with myocardial edema, the regional distribution of
135
LV LGE corresponded to the edematous LV segments in 8 (consistent with an acute myocardial
136
ischemic injury without previous myocardial infarction) and extended to other non-edematous
137
segments in 4 (consistent with an acute myocardial ischemic injury superimposed on a previous
138
post-myocardial infarction scar).
140
Right ventricular LGE was present in a patient with right ventricular myocardial infarction due to occlusion of the proximal right coronary artery.
SC
139
141
No obstructive CAD
M AN U
142
RI PT
133
Among the 26 (59%) patients with non-obstructive CAD, coronary angiography revealed
144
normal coronary arteries in 20 and non-obstructive coronary artery stenosis in 6. A negative history
145
of cardiac disease at risk of sudden cardiac death was documented in all but one patient with a
146
previous diagnosis of HCM. CMR allowed to identify a pathological myocardial substrate in 19
147
(73%) consistent with dilated cardiomyopathy (N=5); acute myocarditis (N=4, Figure 3), mitral
148
valve prolapse associated with LGE (N=3, Figure 4); ischemic heart disease (N=2), isolated non-
149
ischemic left ventricular scar (N=2, Figure 5), arrhythmogenic cardiomyopathy (N=1), hypertrophic
150
cardiomyopathy (N=1) and Tako-Tsubo cardiomyopathy (N=1) (Table 2).
AC C
EP
TE D
143
151
Left ventricular myocardial edema was present in 6 (23%) patients. The following regional
152
distribution patterns of myocardial edema were identified: 1) midmyocardial or subepicardial
153
pattern involving the infero-lateral LV wall in 4 patients, consistent with a diagnosis of acute
154
myocarditis which was confirmed by endomyocardial biopsy in 3; 2) a subendocardial pattern in the
155
mid-apical antero-septal LV segments consistent with myocardial ischemia in one patient with a left
156
anterior descending artery stenosis that was judged as moderate at coronary angiography; 3) a
7
ACCEPTED MANUSCRIPT 157
transmural pattern affecting circumferentially the mid-apical LV segments in one patient who
158
received a diagnosis of Tako-tsubo cardiomyopathy. Left ventricular LGE was present in 18 (69%) patients, including 2 with ischemic
160
distribution (subendocardial or transmural) and 16 with a non-ischemic pattern (midmyocardial or
161
subepicardial). Non-ischemic myocardial fibrosis was found in isolation in 2 patients, and
162
associated with other cardiac diseases in the remaining 15 cases (i.e., cardiomyopathies, N=7; acute
163
myocarditis, N=4; and mitral valve prolapse, N=3).
diagnostic criteria for arrhythmogenic cardiomyopathy[10].
166 167
Incremental diagnostic value of CMR
SC
165
Right ventricular LGE was present in two patients who fulfilled the International Task Force
M AN U
164
RI PT
159
In all 18 patients with obstructive CAD, tissue characterization by CMR confirmed the
169
clinical diagnosis of arrhythmic OHCA due to either acute coronary syndrome (N=12) or chronic
170
ischemic cardiac disease (post-infarction left ventricular scar with no evidence of acute myocardial
171
edema) (N=6).
174 175
diagnosis in 11 (42%). Details are reported in Table 3 and Supplemental file.
EP
173
Among the 26 OHCA survivors with no obstructive CAD, CMR modified the initial clinical
AC C
172
TE D
168
Clinical management
176
An ICD was implanted in all patients with no evidence of acute ischemic injury (no LV
177
myocardial edema), but one who underwent surgical coronary revascularization and had a negative
178
electrophysiological study with programmed ventricular stimulation performed 12 days after
179
surgery. In addition, 5/12 OHCA survivors with evidence of acute myocardial infarction also
180
received an ICD because of severe reduction of LV ejection fraction after 40 days despite optimal
8
ACCEPTED MANUSCRIPT 181
medical therapy (N=4) or moderate LV dysfunction in association with non-sustained VT and
182
inducibility of sustained VT at programmed ventricular stimulation (N=1). Among the 26 OHCA survivors with no obstructive CAD, an ICD was implanted in 22. The
184
remaining 4 OCHA survivors did not receive an ICD for the following reasons: one with mitral
185
valve prolapse and severe mitral regurgitation underwent mitral valve repair and had a negative
186
electrophysiological study with programmed ventricular stimulation 15 days after surgery; one
187
patient with arrhythmogenic cardiomyopathy refused the device implantation; and in 2 patients with
188
acute myocarditis the ICD was deemed not indicated.
SC
189
Follow-up
M AN U
190
RI PT
183
Overall, the mean follow-up period was 36±17 months and no patients were lost to follow-
192
up. Five patients experienced appropriate ICD interventions and one patient, who refused ICD
193
implantation, died suddenly. The 6 patients with arrhythmic events during follow-up showed no
194
myocardial edema at CMR while all 18 patients who received a diagnosis of reversible myocardial
195
damage based on CMR evidence of acute myocardial ischemia (N=13), acute myocarditis (N=4) or
196
Tako-Tsubo cardiomyopathy (N=1), had an uneventful arrhythmic outcome. Five patients with
197
obstructive CAD had severe LV dysfunction at CE-CMR: of those, 3 died for non-arrhythmic
198
causes (refractory heart failure, acute myocardial infarction and abdominal aorta aneurysm rupture)
199
and 1 experienced appropriate ICD intervention. The other 5 patients with arrhythmic events during
200
follow-up had a normal or mildly reduced LV function.
AC C
EP
TE D
191
9
ACCEPTED MANUSCRIPT
DISCUSSION
201
In survivors of OHCA the identification of the underlying arrhythmic myocardial substrate
203
has important implications for patients’ management. Arrhythmic cardiac arrest may be the result of
204
different mechanisms and substrates which include: i) acute and transient electrical instability such
205
that occurring in the setting of acute myocardial ischemia or myocarditis, ii) rapid VT related to a
206
chronic myocardial scar resulting from a previous myocardial infarction or in the context of a
207
cardiomyopathy and iii) primarily electrical disease including genetically-determined ion channel
208
disease, drug toxicity or electrolyte imbalance in the absence of structural heart diseases. With
209
regard to the pathophysiological meaning of myocardial edema, it may either represent the
210
bystander hallmark of an acute cardiac injury (e.g. in acute myocardial infarction) or directly cause
211
ventricular arrhythmias by causing depolarization/repolarization inhomogeneity of myocardium
212
[11].
M AN U
SC
RI PT
202
According to current guidelines[1-2], ICD therapy for secondary prevention of sudden death
214
is reserved to patients with a stable myocardial substrate which exposes to persistent risk of VT/VF
215
relapses. On the contrary, ICD implantation is not justified in patients who suffered VT/VF due to
216
reversible causes, Interpretation of standard investigations such as ECG, echocardiography and
217
coronary angiography may not be straightforward to detect the cause of OHCA and structural
218
substrates of life-threatening ventricular arrhythmias may remain concealed[12-13]; in addition,
219
features of the so-called “post-resuscitation syndrome” may further confuse the clinical diagnosis[3-
220
5].
AC C
EP
TE D
213
221
This study was designed to evaluate the diagnostic yield and prognostic implication of early
222
CMR study with a comprehensive protocol for tissue characterization, including T2-weighted
223
sequences for myocardial edema and post-gadolinium sequences for myocardial fibrosis in a cohort
224
of OHCA survivors with and without obstructive coronary artery disease at coronary angiography 10
ACCEPTED MANUSCRIPT who underwent CE-CMR within one week after the event. The main findings were the following: 1)
226
acute CMR study provided additional value for the identification of causes and mechanisms
227
responsible for arrhythmic OHCA; 2) the evaluation of the presence and regional distribution of
228
both myocardial edema and fibrosis allowed to differentiate acute and reversible myocardial lesions
229
from chronic and stable myocardial substrates, with significant implications on ICD therapy
230
decision-making; 3) identification of acute myocardial edema by CMR predicted an uneventful
231
arrhythmic outcome during follow-up; 4) in our population of OHCA survivors, we did not find any
232
tissue abnormalities at CMR suggestive of post-resuscitation myocardial contusion.
233
Previous studies
M AN U
SC
RI PT
225
Previous studies showed that CMR provides incremental diagnostic value in patients who
235
survived an OHCA [14-17], but CMR scan protocols were often not uniform and performed late
236
after the acute event, thus reducing the sensitivity of T2-weighted sequences for myocardial edema.
237
Moreover, follow-up data to assess the prognostic value of comprehensive CMR tissue
238
characterization in OHCA are lacking. More details are reported in the supplemental file.
TE D
234
Our results confirm and extend previous observations on the diagnostic and prognostic value
240
of CMR in the setting of OHCA. In our study, the tissue characterization imaging protocol
241
incorporated systematic evaluation of both myocardial edema and LGE in order to distinguish acute
242
and potentially reversible myocardial injury from chronic and irreversible tissue damage[6-7]. At
243
variance with previous studies, all our patients underwent early CMR, i.e. within 7 days of
244
hospitalization, in order to provide the highest sensitivity for detection of transient myocardial
245
edema and were followed-up for a mean of 3 years to prospectively assess the prognostic value of
246
CMR findings.
247
Diagnostic yield
AC C
EP
239
11
ACCEPTED MANUSCRIPT In patients with obstructive CAD, CMR confirmed the diagnosis of ischemic heart disease
249
by showing LGE with an ischemic distribution (subendocardial or transmural) in all cases. In
250
addition, T2-weighted sequences allowed to differentiate the arrhythmic substrate due to acute
251
myocardial ischemia, as suggested by the presence of myocardial edema in the territory tributary of
252
a coronary artery, from that of post-infarction myocardial scar[18]. The tissue characterization was
253
also useful for identification or confirmation of the culprit coronary artery, particularly in patients
254
with multiple vessels disease and equivocal coronary angiography findings.
RI PT
248
In patients without obstructive CAD, CMR confirmed the clinical diagnosis in 58% of cases
256
and modified the clinical diagnosis in 42% by demonstrating previously unrecognized myocardial
257
substrates. Myocardial edema at T2-weighted sequences was present in 6 cases suggesting that the
258
OHCA was the result of an acute myocardial injury occurring in the setting of acute coronary
259
syndrome, myocarditis or Tako-tsubo cardiomyopathy. In the remaining patients without
260
myocardial edema, ventricular fibrillation was related to cardiomyopathies or recently identified
261
substrates of SCD such as the isolated non-ischemic myocardial scar [13] or the “arrhythmic” mitral
262
valve prolapse (i.e. floppy mitral valve associated with myocardial fibrosis that is localized at the
263
base of the papillary muscle or at the basal posterior wall behind the mitral leaflet) [19].
264
Prognostic implications
AC C
EP
TE D
M AN U
SC
255
265
Although definite conclusions cannot be drawn because of the limited sample size, our study
266
results suggest that CMR may provide not only diagnostic but also prognostic information.
267
Compared with tissue characterization based on LGE alone, demonstration of myocardial edema by
268
T2-weighted CMR sequences was of additional value for establishing the acute and potentially
269
reversible nature of the arrhythmic myocardial substrate. These results have the potential to impact
270
ICD decision making because indications to device implantation depends on demonstration of
271
chronic/non-reversible myocardial lesion as the substrate underlying VT/VF[1-2]. In our study,
272
patients with myocardial edema suggesting an acute and potentially reversible myocardial lesion 12
ACCEPTED MANUSCRIPT showed no arrhythmic events over a mean follow-up of 3 years. By comparison, 23% of patients
274
without myocardial edema at CMR, who most likely had a chronic and non-reversible substrate,
275
suffered arrhythmic events during follow-up.
276
Myocardial damage secondary to resuscitation
277
RI PT
273
The diagnostic accuracy of CMR has been definitely demonstrated in non-OHCA patients and it is now applied in a variety of cardiac conditions. However, clinical interpretation of tissue
279
abnormalities in survivors of OHCA poses additional challenges because all patients received
280
cardiopulmonary resuscitation that may cause cardiac contusion. In our study we found no signs
281
suggestive of post-resuscitation myocardial contusions. Accordingly, myocardial edema at T2-
282
weighted sequences in patients with OHCA should be interpreted as the sign of an underlying
283
disease rather than a non-specific post-resuscitation injury. More details are reported in the
284
Supplemental File.
285
Study limitations
M AN U
TE D
286
SC
278
The main limitations of the study are that it was retrospective and conducted at a single center. Moreover, our study reported on a relatively small number of patients and outcomes because
288
survivors of OHCA were rare and only a minority of them (32%) were eligible for early CMR.
289
Accordingly, our findings should be confirmed by further studies on larger populations and longer
290
follow-up before the use of early CMR for ICD-decision making in survivors of OHCA can be
291
clinically implemented.
292
AC C
EP
287
In our study, the majority of patients were diagnosed with non-ischemic heart disease in
293
contrast with previous studies demonstrating that obstructive coronary artery disease is the most
294
common cause of OHCA [3]. This finding reflects the selection of patients because only OHCA
295
survivors who were able to undergo early CMR were included in the study. Therefore, the 13
ACCEPTED MANUSCRIPT 296
distribution of OHCA substrates of our study sample cannot be considered representative of the
297
general population.
298
Conclusions
In survivors of OHCA, tissue characterization by early CMR was of significant diagnostic
300
and prognostic value in addition to standard clinical investigation (Supplemental figures 1 and 2).
301
Demonstration of myocardial edema by T2-weighted sequences, suggesting an acute and potentially
302
reversible myocardial injury, was associated with an uneventful arrhythmic follow-up. These
303
findings suggest that early CMR with T2-weighted sequences may be helpful to guide the decision
304
to implant an ICD for secondary prevention in survivors of OHCA that, according to current
305
guidelines, depends on the exclusion of possible transient and reversible causes for VT/VF such as
306
acute myocardial ischemia or acute myocarditis.
M AN U
SC
RI PT
299
TE D
307
Funding Sources: the study was founded by the Target Projects RF-2013-02356762 grant of the
309
Italian Ministry of Health; the TRANSAC (translational arrhythmogenic cardiomyopathy) project,
310
2013, University of Padova, and the research grant BIRD (budget for integrated departmental
311
research) 2016, University of Padova, Italy.
AC C
EP
308
14
ACCEPTED MANUSCRIPT REFERENCES
313
[1] Priori SG, Blomstrom-Lundqvist C. 2015 European Society of Cardiology Guidelines for the
314
management of patients with ventricular arrhythmias and the prevention of sudden cardiac
315
death summarized by co-chairs. Eur Heart J.2015;36:2757-9.
RI PT
312
[2] Al-Khatib SM, Stevenson WG, Ackerman MJ, et al. 2017 AHA/ACC/HRS Guideline for
317
Management of Patients With Ventricular Arrhythmias and the Prevention of Sudden Cardiac
318
Death: Executive Summary: A Report of the American College of Cardiology/American
319
Heart Association Task Force on Clinical Practice Guidelines and the Heart Rhythm Society.
320
Heart Rhythm.2017;doi:10.1016/j.hrthm.2017.10.035.
M AN U
SC
316
321
[3] Anyfantakis ZA, Baron G, Aubry P, Himbert D, Feldman LJ, Juliard JM, Ricard-Hibon A,
322
Burnod A, Cokkinos DV, Steg PG. Acute coronary angiographic findings in survivors of out-
323
of-hospital cardiac arrest. Am Heart J.2009;157:312-8.
325
[4] Price S, Uddin S, Quinn T. Echocardiography in cardiac arrest. Curr Opin Crit Care.2010;16:211-5.
TE D
324
[5] Nolan JP, Neumar RW, Adrie C, et al. Post-cardiac arrest syndrome: epidemiology,
327
pathophysiology, treatment, and prognostication. A Scientific Statement from the
328
International Liaison Committee on Resuscitation; the American Heart Association
329
Emergency Cardiovascular Care Committee; the Council on Cardiovascular Surgery and
330
Anesthesia; the Council on Cardiopulmonary, Perioperative, and Critical Care; the Council on
331
Clinical Cardiology; the Council on Stroke. Resuscitation.2008;79:350-79.
333 334 335
AC C
332
EP
326
[6] Treibel TA, White SK, Moon JC. Myocardial Tissue Characterization: Histological and Pathophysiological Correlation. Curr Cardiovasc Imaging Rep.2014;7:9254. [7] Eitel I, Friedrich MG. T2-weighted cardiovascular magnetic resonance in acute cardiac disease. J Cardiovasc Magn Reson.2011;13:13.
15
ACCEPTED MANUSCRIPT 336
[8] Noc M, Fajadet J, Lassen JF, Kala P, MacCarthy P, Olivecrona GK, Windecker S, Spaulding C.
337
Invasive coronary treatment strategies for out-of-hospital cardiac arrest: a consensus
338
statement from the European association for percutaneous cardiovascular interventions
339
(EAPCI)/stent for life (SFL) groups. EuroIntervention.2014;10:31-7.
341
[9] Moss AJ, Schuger C, Beck CA, et al. Reduction in inappropriate therapy and mortality through
RI PT
340
ICD programming. N Engl J Med.2012;367:2275-83.
[10] Marcus FI, McKenna WJ, Sherrill D, et al. Diagnosis of arrhythmogenic right ventricular
343
cardiomyopathy/dysplasia: proposed modification of the task force criteria. Circulation
344
2010;121:1533-41.
SC
342
[11] Migliore F, Zorzi A, Perazzolo Marra M, Iliceto S, Corrado D. Myocardial edema as a
346
substrate of electrocardiographic abnormalities and life-threatening arrhythmias in reversible
347
ventricular dysfunction of takotsubo cardiomyopathy: Imaging evidence, presumed
348
mechanisms, and implications for therapy. Heart Rhythm.2015;12:1867-77.
M AN U
345
[12] Nucifora G, Muser D, Masci PG, et al. Prevalence and prognostic value of concealed structural
350
abnormalities in patients with apparently idiopathic ventricular arrhythmias of left versus right
351
ventricular
352
Electrophysiol.2014;7:456-62.
a
magnetic
resonance
imaging
study.
Circ
Arrhythm
EP
origin:
TE D
349
[13] Zorzi A, Perazzolo Marra M, Rigato I, et al. Nonischemic Left Ventricular Scar as a Substrate
354
of Life-Threatening Ventricular Arrhythmias and Sudden Cardiac Death in Competitive
355
Athletes. Circ Arrhythm Electrophysiol. 2016;9.pii:e004229.
AC C
353
356
[14] White JA, Fine NM, Gula L, Yee R, Skanes A, Klein G, Leong-Sit P, Warren H, Thompson T,
357
Drangova M, Krahn A. Utility of cardiovascular magnetic resonance in identifying substrate
358
for malignant ventricular arrhythmias. Circ Cardiovasc Imaging.2012;5:12-20.
359
[15] Baritussio A, Zorzi A, Ghosh Dastidar A, et al. Out of hospital cardiac arrest survivors with
360
inconclusive coronary angiogram: Impact of cardiovascular magnetic resonance on clinical
361
management and decision-making. Resuscitation.2017;116:91-97. 16
ACCEPTED MANUSCRIPT 362
[16] Marstrand P, Axelsson A, Thune JJ, Vejlstrup N, Bundgaard H, Theilade J. Cardiac magnetic
363
resonance imaging after ventricular tachyarrhythmias increases diagnostic precision and
364
reduces the need for family screening for inherited cardiac disease. Europace.2016;18:1860-5.
365
[17] Neilan TG, Farhad H, Mayrhofer T, et al. Late gadolinium enhancement among survivors of
368 369 370
RI PT
367
sudden cardiac arrest. JACC Cardiovasc Imaging.2015;8:414-23.
[18] Perazzolo Marra M, Lima JA, Iliceto S. MRI in acute myocardial infarction. Eur Heart J. 2011;32:284-93.
[19] Basso C, Perazzolo Marra M, Rizzo S, et al. Arrhythmic mitral valve prolapse and sudden cardiac death. Circulation 2016.132:556-66.
AC C
EP
TE D
M AN U
371
SC
366
17
ACCEPTED MANUSCRIPT 372
FIGURES LEGEND
373
Figure 1. Out-of-hospital cardiac arrest in a patient with obstructive coronary artery disease.
375
Out-of-hospital cardiac arrest occurring in a 64-year-old male with acute coronary syndrome.
376
Coronary angiography showing multiple vessels obstructive (≥70%) coronary artery stenosis
377
(arrows) affecting both the right (A) and the left (B) coronary artery. Cardiac magnetic resonance
378
showing transmural myocardial edema (three chambers, long axis view; T2-weighted sequences)
379
(C) and subendocardial late gadolinium enhancement (three chamber, long axis view; T1-weighted
380
inversion recovery post-contrast sequences) (D) affecting the infero-lateral left ventricular wall
381
which is tributary of the left circumflex coronary artery.
M AN U
SC
RI PT
374
382
Figure 2. Out-of-hospital cardiac arrest in a patient with obstructive coronary artery disease.
384
Out-of-hospital cardiac arrest occurring in a 57-year-old male with post-myocardial infarction scar.
385
Coronary angiography showing multiple vessels obstructive (≥70%) coronary artery stenosis
386
(arrows) affecting both the right (A) and the left (B) coronary artery. Cardiac magnetic resonance
387
showing no myocardial edema (three chambers, long axis view; T2-weighted sequences) (C) and
388
transmural late gadolinium enhancement (three chambers, long axis view; T1-weighted inversion
389
recovery post-contrast sequences) (D) affecting the entire left ventricular lateral wall which is
390
tributary of the left circumflex coronary artery.
AC C
EP
TE D
383
391 392 393
Figure 3. Out-of-hospital cardiac arrest in a patient with no obstructive coronary artery
394
disease.
395
Out-of-hospital cardiac arrest occurring in a 41-year-old male with biopsy-proven acute
396
myocarditis. Coronary angiography showing normal right (A) and left (B) coronary artery. Cardiac 18
ACCEPTED MANUSCRIPT 397
magnetic resonance showing myocardial edema (two chambers, long axis view; T2-weighted
398
sequences) (C) and transmural late gadolinium enhancement (two chambers, long axis view; T1-
399
weighted inversion recovery post-contrast sequences) (D) affecting the mid-apical inferior left
400
ventricular wall.
RI PT
401
Figure 4. Out-of-hospital cardiac arrest in a patient with no obstructive coronary artery
403
disease.
404
Out-of-hospital cardiac arrest occurring in a 29-year-old female with arrhythmic mitral valve
405
prolapse. Coronary angiography showing normal right (A) and left (B) coronary artery. Cardiac
406
magnetic resonance showing thickening and prolapse of both mitral valve leaflets (C,D), no
407
myocardial edema (three chambers, long axis view; T2-weighted sequences) (C) and evidence of
408
mid-myocardial late gadolinium enhancement (three chambers, long axis view; T1-weighted
409
inversion recovery post-contrast sequences) (D) affecting the basal infero-lateral left ventricular
410
wall behind the prolapsing posterior mitral valve leaflet.
411
TE D
M AN U
SC
402
Figure 5. Out-of-hospital cardiac arrest in a patient with no obstructive coronary artery
413
disease.
414
Out-of-hospital cardiac arrest occurring in a 36-year-old male with an isolated non-ischemic left
415
ventricular scar. Coronary angiography showing normal right (A) and left (B) coronary artery.
416
Cardiac magnetic resonance showing no myocardial edema (four chambers, long axis view; T2-
417
weighted sequences) (C) and the presence of subepicardial late gadolinium enhancement (four
418
chambers, long axis view; T1-weighted inversion recovery post-contrast sequences) (D) affecting
419
almost the entire left ventricular lateral wall.
AC C
EP
412
420 421
19
ACCEPTED MANUSCRIPT TABLES
422
Table 1: clinical characteristics of the study population Coronary stenosis ≥ 70% N=18 17 (94%) 57 (54-71)
NO Coronary stenosis ≥ 70% N=26 19 (73%) 44 (27-59)
P
0.11 Gender (male) <0.001 Age (years) Risk factors Hypertension 13 (72%) 5 (19%) 0.001 Diabetes 6 (33%) 2 (8%) 0.05 Smoking 8 (45%) 4 (15%) 0.04 Dyslipidemia 8 (44%) 5 (19%) 0.09 Family history for CAD 4 (22%) 4 (15%) 0.70 Personal history Ischemic heart disease 4 (22%) 0 0.03 Non-ischemic heart disease 0 3 (12%) 0.25 Chest pain within 1 month 6 (33%) 1 (4%) 0.02 Syncope/pre-syncope 0 3 (12%) 0.25 Characteristics of cardiac arrest Cardiac arrest in public place 11 (61%) 9 (35%) 0.06 Witnessed cardiac arrest 17 (95%) 23 (88%) 0.63 Bystander CPR 11 (61%) 22 (85%) 0.09 ST-segment elevation post 7 (39%) 0 <0.001 resuscitation Echocardiography at admission 42 (33-50) 50 (47-56) 0.014 LV EF (%) 69 (60-84) 0.40 71 (56-96) LV EDVi (ml/m2) <0.001 LV regional WMA 17 (94%) 8 (31%) 2 (11%) 5 (19%) 0.68 RV dilation 2 (8%) 0.68 RV dysfunction 2 (11%) RV regional WMA 1 (6%) 3 (12%) 1.0 Mitral valve regurgitation ≥ 2 (11%) 3 (12%) 1.0 moderate 5 (19%) 0.06 Mitral valve prolapse 0 Aortic valve stenosis ≥ moderate 0 0 5.1 (1.5-82) 1.2 (0.29-9.9) 0.02 Troponin I peak (umol/L) Coronary angiography At admission 12 (67%) 8 (31%) 0.41 Time from admission 1 (0-3) 2 (1-3) 0.55 Single vessel disease 5 (28%) 0 Multiple vessel disease with 5 (28%) 0 culprit lesion Clinical diagnosis (no CMR) Structurally normal heart* 0 12 (46%) <0.001 Ischemic heart disease 18 (100%) 1 (4%) Acute myocarditis 0 3 (12%) Cardiomyopathy# 0 5 (19%) Mitral valve prolapse 0 5 (19%) * including Brugada syndrome, idiopathic catecholaminergic ventricular tachycardia, long QT syndrome, and short QT syndrome # including non-ischemic dilated cardiomyopathy, arrhythmogenic cardiomyopathy and hypertrophic cardiomyopathy
AC C
EP
TE D
M AN U
SC
RI PT
423
424 425 426 427 428 429 430
CMR=cardiac magnetic resonance; ICD=implantable cardioverter defibrillator; LV=left ventricular; RV=right ventricular; TAPSE=tricuspid annular plane systolic excursion; WMA= wall motion abnormalities; TTS=Tako-tsubo syndrome
20
ACCEPTED MANUSCRIPT Table 2: cardiac magnetic resonance findings in the study population P
3 (17%) 3 (17%) 7 (39%) 5 (28%)
18 (69%) 6 (24%) 2 (8%) 0
<0.001
7 (39%) 6 (33%) 2 (11%) 3 (17%) 17 (94%) 1 (6%) 0 2 (11%)
20 (77%) 3 (12%) 2 (8%) 1 (4%) 5 (19%) 4 (15%) 1 (4%) 3 (12%)
6 (33%) 0
20 (77%) 4 (15%)
1 (6%) 11 (61%)
1 (4%) 1 (4%)
0 0 7 (39%) 11 (61%) 1 (6%)
7 (27%) 17 (65%) 1 (4%) 1 (4%) 2 (8%)
<0.001
7 (27%) 2 (8%) 4 (15%) 7 (27%) 3 (12%) 1 (4%) 2 (8%)
<0.001
RI PT
NO Coronary stenosis ≥ 70% N=26
0.03
<0.001 0.52 1.0 1.0
SC
M AN U
0 18 (100%) 0 0 0 0 0
EP
Morpho-functional parameters LV systolic function Normal Mild impairment Moderate impairment Severe impairment LVEDV Normal Mild dilation Moderate dilation Severe dilation LV regional WMA RV systolic impairment RV dilatation RV regional WMA LV Myocardial edema Absent Patchy, midmyocardial or subepicardial Subendocardial Transmural LV LGE Absent Midmyocardial or subepicardial Subendocardial Transmural RV LGE CMR diagnosis Structurally normal heart* Ischemic heart disease Acute myocarditis Cardiomyopathy# Mitral valve prolapse with LGE Tako-Tsubo syndrome Non-ischemic LV scar
Coronary stenosis ≥ 70% N=18
TE D
431
0.002
1.0
CMR=cardiac magnetic resonance; LGE=late-gadolinium enhancement; LVEDV= left ventricular end-diastolic volume; LV=left ventricular; RV=right ventricular; WMA= wall motion abnormalities;
436
# including non-ischemic dilated cardiomyopathy, arrhythmogenic cardiomyopathy and hypertrophic cardiomyopathy
437
AC C
432 433 434 435
21
ACCEPTED MANUSCRIPT 438
Table 3 – comparison between clinical and cardiac magnetic resonance diagnosis Suspected diagnosis (pre-CMR) Structurally normal heart (N=12)
Acute myocarditis (N=3)
Cardiomyopathy (N=5)
SC
Mitral valve prolapse (N=5)
M AN U
Agreement
18/19 (95%) 1/3 (33%)
4/5 (80%)
3/5 (60%)
18/18 (100%) 15/26 (58%) 33/44 (75%)
EP
TE D
CMR=cardiac magnetic resonance; LGE=late gadolinium enhancement; LV=left ventricular
AC C
439
Agreement 7/12 (58%)
RI PT
Ischemic heart disease (N=19)
Final diagnosis (after-CMR) Structurally normal heart = 7 Acute myocarditis = 2 Cardiomyopathy = 1 Non-ischemic LV scar = 2 Ischemic heart disease = 18 Tako-tsubo syndrome = 1 Acute myocarditis = 1 Ischemic heart disease = 1 Cardiomyopathy = 1 Cardiomyopathy = 4 Ischemic heart disease = 1 Mitral valve prolapse with LGE = 3 Cardiomyopathy = 1 Acute myocarditis = 1 Coronary stenosis ≥70% group NO coronary stenosis ≥70% group Overall agreement
22
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S1547-5271(18)30195-4
DOI:
10.1016/j.hrthm.2018.02.033
Reference:
HRTHM 7502
To appear in:
Heart Rhythm
Received Date: 12 December 2017
Please cite this article as: Zorzi A, Susana A, De Lazzari M, Migliore F, Vescovo G, Scarpa D, Baritussio A, Tarantini G, Cacciavillani L, Giorgi B, Basso C, Iliceto S, Bucciarelli Ducci C, Corrado D, Marra MP, Diagnostic value and prognostic implications of early cardiac magnetic resonance in survivors of out of hospital cardiac arrest, Heart Rhythm (2018), doi: 10.1016/j.hrthm.2018.02.033. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Diagnostic value and prognostic implications of early cardiac magnetic
2
resonance in survivors of out of hospital cardiac arrest
3
Alessandro Zorzi, MD, PhD (1); Angela Susana, MD (1); Manuel De Lazzari, MD, PhD (1);
4
Federico Migliore, MD, PhD (1); Giovanni Vescovo, MD (1); Daniele Scarpa, MD (1); Anna
5
Baritussio, MD (1,3); Giuseppe Tarantini, MD, PhD (1); Luisa Cacciavillani, MD, PhD (1);
6
Benedetta Giorgi, MD (2); Cristina Basso, MD, PhD (1); Sabino Iliceto, MD (1); Chiara Bucciarelli
7
Ducci, MD, PhD (3); Domenico Corrado, MD, PhD (1); Martina Perazzolo Marra, MD, PhD (1).
SC
8
RI PT
1
1) Division of cardiology Department of Cardiac, Thoracic and Vascular Sciences, University of Padova, Italy
M AN U
9 10
2) Division of radiology, Department of Medicine, Az. Ospedaliera di Padova, Italy
11
3) Bristol NIHR Cardiovascular Biomedical Research Unit, Bristol Heart Institute, University of Bristol, United Kingdom
12 13
TE D
14 15 16
Correspondence to:
Domenico Corrado, MD, PhD
18
Department of Cardiac, Thoracic and Vascular Sciences
19
University of Padova
20
Via Giustiniani, 2 - 35121 Padova (Italy)
21
Tel. +39 049 8212458; Fax +39 049 8212309
22
e-mail: [email protected]
AC C
EP
17
23 24
Short title: cardiac magnetic resonance findings in survivors of out-of-hospital cardiac arrest
25
Word count (including title page, abstract, text, references, tables and figures legend): 5065
26
Conflicts of interest: none
1
ACCEPTED MANUSCRIPT
ABSTRACT
28
Background: In patients who survived out-of-hospital cardiac arrest (OHCA) it is crucial to
29
establish the underlying cause and its potential reversibility. Objective: We assessed the
30
incremental diagnostic and prognostic role of early cardiac magnetic resonance (CMR) in survivors
31
of OHCA. Methods: Among 139 consecutive OHCA patients, we enrolled 44 (median age 43
32
years; 84% males) patients who underwent coronary angiography and CMR ≤7 days after
33
admission. The CMR protocol included T2-weighted sequences for myocardial edema and late
34
gadolinium-enhancement (LGE) sequences for myocardial fibrosis. Results: Coronary angiography
35
identified obstructive coronary artery disease (CAD) in 18/44 patients, in whom CMR confirmed
36
the diagnosis of ischemic heart disease by demonstrating subendocardial or transmural LGE; the
37
presence of myocardial edema allowed to differentiate between acute myocardial ischemia (N=12)
38
and post-infarction myocardial scar (N=6). Among the remaining 26 patients without obstructive
39
CAD, 19 (73%) showed at CMR dilated cardiomyopathy (N=5), myocarditis (N=4), mitral valve
40
prolapse associated with LGE (N=3), ischemic scar (N=2), idiopathic non-ischemic scar (N=2),
41
arrhythmogenic cardiomyopathy (N=1), hypertrophic cardiomyopathy (N=1) and Tako-Tsubo
42
cardiomyopathy (N=1). In this subgroup, 6/26 (23%) had myocardial edema. During a mean
43
follow-up of 36±17 months, all 18 patients with myocardial edema had an uneventful outcome
44
while 9/26 (35%) without myocardial edema experienced sudden arrhythmic death (N=1),
45
appropriate defibrillator interventions (N=5) and non-arrhythmic death (N=3) (p=0.006).
46
Conclusions: In survivors of OHCA, early CMR with a comprehensive tissue characterization
47
protocol provided additional diagnostic and prognostic value. The identification of myocardial
48
edema was associated with a favorable long-term outcome.
49
Key Words: cardiac arrest; cardiopulmonary resuscitation; cardiac blunt trauma; cardiac magnetic
50
resonance; implantable cardioverter defibrillator; secondary prevention; ventricular arrhythmia.
AC C
EP
TE D
M AN U
SC
RI PT
27
2
ACCEPTED MANUSCRIPT INTRODUCTION
52
Out-of-hospital cardiac arrest (OHCA) due to ventricular fibrillation (VF) or pulseless
53
ventricular tachycardia (VT) can be caused by ischemic and non-ischemic heart diseases. In patients
54
who suffered OHCA, clinical investigations aim at establishing what is the underlying disease and
55
whether the cause of the arrhythmia is reversible. This is particularly relevant for decision to
56
implant a defibrillator (ICD), that is not indicated in patients who experienced VT/VF due to a
57
reversible myocardial injury like that secondary to myocardial ischemia of acute coronary
58
syndromes or myocarditis[1-2]. In this contest, the interpretation of standard investigations such as
59
ECG, echocardiography and coronary angiography may not be straightforward and nonspecific
60
findings secondary to the cardiac arrest and subsequent resuscitation may act as confounders[3-5].
M AN U
SC
RI PT
51
Cardiac magnetic resonance (CMR), in addition to accurate assessment of ventricular
62
function, allows the identification of myocardial tissue changes such as myocardial edema as
63
evidenced by increased signal intensity on T2-weighted sequences and myocardial fibrosis by late
64
gadolinium enhancement (LGE) on post-contrast sequences[6]. In the evaluation of patients who
65
survived OHCA, the combination of sequences for myocardial edema and fibrosis has the potential
66
to distinguish an acute and potentially reversible injury from a chronic and irreversible lesion[7].
67
The aim of the present study was to prospectively evaluate the incremental diagnostic and
AC C
EP
TE D
61
68
prognostic value of early (within one week after admission) CMR, enhanced with an accurate tissue
69
characterization protocol including sequences for myocardial edema and fibrosis, in OHCA
70
survivors.
3
ACCEPTED MANUSCRIPT METHODS
71
72
The study included all survivors of arrhythmic OHCA who were admitted to our institution during the study period 2011-2016 and underwent coronary angiography and CMR within one week
74
of hospitalization. The CMR was performed in all patients who could be weaned from mechanical
75
ventilation, had no neurological impairment and had no contraindications to CMR. The study
76
complies with the Declaration of Helsinki, was approved by the Ethical Committee and all patients
77
agreed to participate.
78
Coronary angiography
SC
RI PT
73
Coronary angiography was performed in all patients: at admission in those with ST-segment
80
elevation at post-resuscitation ECG or in those with high clinical suspicious of ongoing myocardial
81
ischemia; delayed in the others. A culprit coronary lesion was defined as obstructive (≥70%) CAD
82
with TIMI 0/1 flow with abrupt closure, or TIMI 2/3 flow with features suggestive of thrombus or
83
ulcerated plaques, ST segment and T wave changes in the corresponding ECG location, and
84
evidence of matching regional wall motion abnormality on left ventriculogram or
85
echocardiogram[8].
86
Contrast-enhanced cardiac magnetic resonance
TE D
EP
The CMR scan protocol systematically included steady-state free precession sequence (true
AC C
87
M AN U
79
88
FISP) cine images for morpho-functional evaluation, T2-weighted sequences for detection of
89
myocardial edema and post-contrast sequences for detection of LGE. Details are reported in the
90
supplemental file.
91
Follow-up
92
Patients were followed-up with serial outpatient evaluations or with telephonic interviews to
93
determine the alive status and whether they reached the composite arrhythmic end-point including
94
sudden death, cardiac arrest due to VF, sustained VT and appropriate ICD intervention. Sudden 4
ACCEPTED MANUSCRIPT death was defined as any natural death occurring instantaneously or within 60 minutes from
96
symptoms onset. Appropriate ICD intervention was defined as an ICD shock delivered in response
97
to a ventricular tachyarrhythmia >170 bpm and documented by stored intracardiac ECG data. All
98
ICD interventions were adjudicated as appropriate or inappropriate by two electrophysiologists
99
(AZ, DC) based on morphologic features, tachycardia onset, rate stability and atrial electrograms
100
(when available). ICD were programmed according to the MADIT-RIT trial high-rate or delayed-
101
therapy programming[9].
RI PT
95
104
Statistical Analysis
M AN U
103
SC
102
Continuous and categorical variables are expressed as median (with 25th-75th %iles), and n (%), respectively. Because of the small sample size, normality distribution was not assumed.
106
Categorical variables were compared by using the chi-square or Fisher exact test, as appropriate.
107
Continuous data were compared using the Mann-Whitney U test. A p-value of <0.05 was
108
considered statistically significant. Data were analysed with SPSS® version 23 (IBM®).
AC C
EP
TE D
105
5
ACCEPTED MANUSCRIPT
RESULTS
109
Among a consecutive series of 139 patients admitted for OHCA, 37 (27%) died during the
111
acute phase. Among the remaining 102 patients, 58 (42%) were not recruited because of
112
neurological impairment and/or mechanical ventilation (N=52), the presence of a no-MRI
113
compatible pacemaker (N=4) or poor CMR quality images (N=2). Forty-four patients [median age
114
43 years (39-62 years); 84% males] met the enrollment criteria and constituted the definitive study
115
cohort. (Table 1). Coronary angiography was performed at admission in 20 (45%) patients,
116
including 7 (16%) with post-resuscitation ST-segment elevation, and within 4 days in the remaining
117
patients.
M AN U
SC
RI PT
110
118 119
Obstructive CAD
Angiographic evidence of obstructive coronary disease (≥1 coronary stenosis ≥70%) was
121
found in 18 (41%) patients. A clear culprit coronary lesion was identified in 10 patients, 5/5 with
122
single vessel disease and 5/13 with multiple vessels disease.
TE D
120
Cardiac magnetic resonance findings are summarized in Table 2.
124
The presence of myocardial edema at T2-weighted sequences consistent with an acute
125
ischemic injury in 12/18 (67%) patients (Figure 1). In all cases, the regional distribution of
126
myocardial edema was concordant with the territory which was tributary of a coronary artery with
127
obstructive stenosis or occlusion. In 10 patients with an identifiable culprit coronary lesion, the
128
regions of myocardial edema were always concordant with the regional distribution of the culprit
129
coronary artery. Myocardial edema was present in 2 of the 8 patients with multiple vessels disease
130
and no identifiable culprit coronary lesion. The Peak troponin I (normal value <0.017 ug/L) was
131
significantly higher in patients with [12.2 ug/L (2.1-104.8)] compared with those without [2.8 ug/L
132
(1.2-26.1)] LV myocardial edema (p = 0.02).
AC C
EP
123
6
ACCEPTED MANUSCRIPT Left ventricular LGE with a subendocardial or transmural distribution was demonstrated in
134
all patients (Figure 2). Among the 12 patients with myocardial edema, the regional distribution of
135
LV LGE corresponded to the edematous LV segments in 8 (consistent with an acute myocardial
136
ischemic injury without previous myocardial infarction) and extended to other non-edematous
137
segments in 4 (consistent with an acute myocardial ischemic injury superimposed on a previous
138
post-myocardial infarction scar).
140
Right ventricular LGE was present in a patient with right ventricular myocardial infarction due to occlusion of the proximal right coronary artery.
SC
139
141
No obstructive CAD
M AN U
142
RI PT
133
Among the 26 (59%) patients with non-obstructive CAD, coronary angiography revealed
144
normal coronary arteries in 20 and non-obstructive coronary artery stenosis in 6. A negative history
145
of cardiac disease at risk of sudden cardiac death was documented in all but one patient with a
146
previous diagnosis of HCM. CMR allowed to identify a pathological myocardial substrate in 19
147
(73%) consistent with dilated cardiomyopathy (N=5); acute myocarditis (N=4, Figure 3), mitral
148
valve prolapse associated with LGE (N=3, Figure 4); ischemic heart disease (N=2), isolated non-
149
ischemic left ventricular scar (N=2, Figure 5), arrhythmogenic cardiomyopathy (N=1), hypertrophic
150
cardiomyopathy (N=1) and Tako-Tsubo cardiomyopathy (N=1) (Table 2).
AC C
EP
TE D
143
151
Left ventricular myocardial edema was present in 6 (23%) patients. The following regional
152
distribution patterns of myocardial edema were identified: 1) midmyocardial or subepicardial
153
pattern involving the infero-lateral LV wall in 4 patients, consistent with a diagnosis of acute
154
myocarditis which was confirmed by endomyocardial biopsy in 3; 2) a subendocardial pattern in the
155
mid-apical antero-septal LV segments consistent with myocardial ischemia in one patient with a left
156
anterior descending artery stenosis that was judged as moderate at coronary angiography; 3) a
7
ACCEPTED MANUSCRIPT 157
transmural pattern affecting circumferentially the mid-apical LV segments in one patient who
158
received a diagnosis of Tako-tsubo cardiomyopathy. Left ventricular LGE was present in 18 (69%) patients, including 2 with ischemic
160
distribution (subendocardial or transmural) and 16 with a non-ischemic pattern (midmyocardial or
161
subepicardial). Non-ischemic myocardial fibrosis was found in isolation in 2 patients, and
162
associated with other cardiac diseases in the remaining 15 cases (i.e., cardiomyopathies, N=7; acute
163
myocarditis, N=4; and mitral valve prolapse, N=3).
diagnostic criteria for arrhythmogenic cardiomyopathy[10].
166 167
Incremental diagnostic value of CMR
SC
165
Right ventricular LGE was present in two patients who fulfilled the International Task Force
M AN U
164
RI PT
159
In all 18 patients with obstructive CAD, tissue characterization by CMR confirmed the
169
clinical diagnosis of arrhythmic OHCA due to either acute coronary syndrome (N=12) or chronic
170
ischemic cardiac disease (post-infarction left ventricular scar with no evidence of acute myocardial
171
edema) (N=6).
174 175
diagnosis in 11 (42%). Details are reported in Table 3 and Supplemental file.
EP
173
Among the 26 OHCA survivors with no obstructive CAD, CMR modified the initial clinical
AC C
172
TE D
168
Clinical management
176
An ICD was implanted in all patients with no evidence of acute ischemic injury (no LV
177
myocardial edema), but one who underwent surgical coronary revascularization and had a negative
178
electrophysiological study with programmed ventricular stimulation performed 12 days after
179
surgery. In addition, 5/12 OHCA survivors with evidence of acute myocardial infarction also
180
received an ICD because of severe reduction of LV ejection fraction after 40 days despite optimal
8
ACCEPTED MANUSCRIPT 181
medical therapy (N=4) or moderate LV dysfunction in association with non-sustained VT and
182
inducibility of sustained VT at programmed ventricular stimulation (N=1). Among the 26 OHCA survivors with no obstructive CAD, an ICD was implanted in 22. The
184
remaining 4 OCHA survivors did not receive an ICD for the following reasons: one with mitral
185
valve prolapse and severe mitral regurgitation underwent mitral valve repair and had a negative
186
electrophysiological study with programmed ventricular stimulation 15 days after surgery; one
187
patient with arrhythmogenic cardiomyopathy refused the device implantation; and in 2 patients with
188
acute myocarditis the ICD was deemed not indicated.
SC
189
Follow-up
M AN U
190
RI PT
183
Overall, the mean follow-up period was 36±17 months and no patients were lost to follow-
192
up. Five patients experienced appropriate ICD interventions and one patient, who refused ICD
193
implantation, died suddenly. The 6 patients with arrhythmic events during follow-up showed no
194
myocardial edema at CMR while all 18 patients who received a diagnosis of reversible myocardial
195
damage based on CMR evidence of acute myocardial ischemia (N=13), acute myocarditis (N=4) or
196
Tako-Tsubo cardiomyopathy (N=1), had an uneventful arrhythmic outcome. Five patients with
197
obstructive CAD had severe LV dysfunction at CE-CMR: of those, 3 died for non-arrhythmic
198
causes (refractory heart failure, acute myocardial infarction and abdominal aorta aneurysm rupture)
199
and 1 experienced appropriate ICD intervention. The other 5 patients with arrhythmic events during
200
follow-up had a normal or mildly reduced LV function.
AC C
EP
TE D
191
9
ACCEPTED MANUSCRIPT
DISCUSSION
201
In survivors of OHCA the identification of the underlying arrhythmic myocardial substrate
203
has important implications for patients’ management. Arrhythmic cardiac arrest may be the result of
204
different mechanisms and substrates which include: i) acute and transient electrical instability such
205
that occurring in the setting of acute myocardial ischemia or myocarditis, ii) rapid VT related to a
206
chronic myocardial scar resulting from a previous myocardial infarction or in the context of a
207
cardiomyopathy and iii) primarily electrical disease including genetically-determined ion channel
208
disease, drug toxicity or electrolyte imbalance in the absence of structural heart diseases. With
209
regard to the pathophysiological meaning of myocardial edema, it may either represent the
210
bystander hallmark of an acute cardiac injury (e.g. in acute myocardial infarction) or directly cause
211
ventricular arrhythmias by causing depolarization/repolarization inhomogeneity of myocardium
212
[11].
M AN U
SC
RI PT
202
According to current guidelines[1-2], ICD therapy for secondary prevention of sudden death
214
is reserved to patients with a stable myocardial substrate which exposes to persistent risk of VT/VF
215
relapses. On the contrary, ICD implantation is not justified in patients who suffered VT/VF due to
216
reversible causes, Interpretation of standard investigations such as ECG, echocardiography and
217
coronary angiography may not be straightforward to detect the cause of OHCA and structural
218
substrates of life-threatening ventricular arrhythmias may remain concealed[12-13]; in addition,
219
features of the so-called “post-resuscitation syndrome” may further confuse the clinical diagnosis[3-
220
5].
AC C
EP
TE D
213
221
This study was designed to evaluate the diagnostic yield and prognostic implication of early
222
CMR study with a comprehensive protocol for tissue characterization, including T2-weighted
223
sequences for myocardial edema and post-gadolinium sequences for myocardial fibrosis in a cohort
224
of OHCA survivors with and without obstructive coronary artery disease at coronary angiography 10
ACCEPTED MANUSCRIPT who underwent CE-CMR within one week after the event. The main findings were the following: 1)
226
acute CMR study provided additional value for the identification of causes and mechanisms
227
responsible for arrhythmic OHCA; 2) the evaluation of the presence and regional distribution of
228
both myocardial edema and fibrosis allowed to differentiate acute and reversible myocardial lesions
229
from chronic and stable myocardial substrates, with significant implications on ICD therapy
230
decision-making; 3) identification of acute myocardial edema by CMR predicted an uneventful
231
arrhythmic outcome during follow-up; 4) in our population of OHCA survivors, we did not find any
232
tissue abnormalities at CMR suggestive of post-resuscitation myocardial contusion.
233
Previous studies
M AN U
SC
RI PT
225
Previous studies showed that CMR provides incremental diagnostic value in patients who
235
survived an OHCA [14-17], but CMR scan protocols were often not uniform and performed late
236
after the acute event, thus reducing the sensitivity of T2-weighted sequences for myocardial edema.
237
Moreover, follow-up data to assess the prognostic value of comprehensive CMR tissue
238
characterization in OHCA are lacking. More details are reported in the supplemental file.
TE D
234
Our results confirm and extend previous observations on the diagnostic and prognostic value
240
of CMR in the setting of OHCA. In our study, the tissue characterization imaging protocol
241
incorporated systematic evaluation of both myocardial edema and LGE in order to distinguish acute
242
and potentially reversible myocardial injury from chronic and irreversible tissue damage[6-7]. At
243
variance with previous studies, all our patients underwent early CMR, i.e. within 7 days of
244
hospitalization, in order to provide the highest sensitivity for detection of transient myocardial
245
edema and were followed-up for a mean of 3 years to prospectively assess the prognostic value of
246
CMR findings.
247
Diagnostic yield
AC C
EP
239
11
ACCEPTED MANUSCRIPT In patients with obstructive CAD, CMR confirmed the diagnosis of ischemic heart disease
249
by showing LGE with an ischemic distribution (subendocardial or transmural) in all cases. In
250
addition, T2-weighted sequences allowed to differentiate the arrhythmic substrate due to acute
251
myocardial ischemia, as suggested by the presence of myocardial edema in the territory tributary of
252
a coronary artery, from that of post-infarction myocardial scar[18]. The tissue characterization was
253
also useful for identification or confirmation of the culprit coronary artery, particularly in patients
254
with multiple vessels disease and equivocal coronary angiography findings.
RI PT
248
In patients without obstructive CAD, CMR confirmed the clinical diagnosis in 58% of cases
256
and modified the clinical diagnosis in 42% by demonstrating previously unrecognized myocardial
257
substrates. Myocardial edema at T2-weighted sequences was present in 6 cases suggesting that the
258
OHCA was the result of an acute myocardial injury occurring in the setting of acute coronary
259
syndrome, myocarditis or Tako-tsubo cardiomyopathy. In the remaining patients without
260
myocardial edema, ventricular fibrillation was related to cardiomyopathies or recently identified
261
substrates of SCD such as the isolated non-ischemic myocardial scar [13] or the “arrhythmic” mitral
262
valve prolapse (i.e. floppy mitral valve associated with myocardial fibrosis that is localized at the
263
base of the papillary muscle or at the basal posterior wall behind the mitral leaflet) [19].
264
Prognostic implications
AC C
EP
TE D
M AN U
SC
255
265
Although definite conclusions cannot be drawn because of the limited sample size, our study
266
results suggest that CMR may provide not only diagnostic but also prognostic information.
267
Compared with tissue characterization based on LGE alone, demonstration of myocardial edema by
268
T2-weighted CMR sequences was of additional value for establishing the acute and potentially
269
reversible nature of the arrhythmic myocardial substrate. These results have the potential to impact
270
ICD decision making because indications to device implantation depends on demonstration of
271
chronic/non-reversible myocardial lesion as the substrate underlying VT/VF[1-2]. In our study,
272
patients with myocardial edema suggesting an acute and potentially reversible myocardial lesion 12
ACCEPTED MANUSCRIPT showed no arrhythmic events over a mean follow-up of 3 years. By comparison, 23% of patients
274
without myocardial edema at CMR, who most likely had a chronic and non-reversible substrate,
275
suffered arrhythmic events during follow-up.
276
Myocardial damage secondary to resuscitation
277
RI PT
273
The diagnostic accuracy of CMR has been definitely demonstrated in non-OHCA patients and it is now applied in a variety of cardiac conditions. However, clinical interpretation of tissue
279
abnormalities in survivors of OHCA poses additional challenges because all patients received
280
cardiopulmonary resuscitation that may cause cardiac contusion. In our study we found no signs
281
suggestive of post-resuscitation myocardial contusions. Accordingly, myocardial edema at T2-
282
weighted sequences in patients with OHCA should be interpreted as the sign of an underlying
283
disease rather than a non-specific post-resuscitation injury. More details are reported in the
284
Supplemental File.
285
Study limitations
M AN U
TE D
286
SC
278
The main limitations of the study are that it was retrospective and conducted at a single center. Moreover, our study reported on a relatively small number of patients and outcomes because
288
survivors of OHCA were rare and only a minority of them (32%) were eligible for early CMR.
289
Accordingly, our findings should be confirmed by further studies on larger populations and longer
290
follow-up before the use of early CMR for ICD-decision making in survivors of OHCA can be
291
clinically implemented.
292
AC C
EP
287
In our study, the majority of patients were diagnosed with non-ischemic heart disease in
293
contrast with previous studies demonstrating that obstructive coronary artery disease is the most
294
common cause of OHCA [3]. This finding reflects the selection of patients because only OHCA
295
survivors who were able to undergo early CMR were included in the study. Therefore, the 13
ACCEPTED MANUSCRIPT 296
distribution of OHCA substrates of our study sample cannot be considered representative of the
297
general population.
298
Conclusions
In survivors of OHCA, tissue characterization by early CMR was of significant diagnostic
300
and prognostic value in addition to standard clinical investigation (Supplemental figures 1 and 2).
301
Demonstration of myocardial edema by T2-weighted sequences, suggesting an acute and potentially
302
reversible myocardial injury, was associated with an uneventful arrhythmic follow-up. These
303
findings suggest that early CMR with T2-weighted sequences may be helpful to guide the decision
304
to implant an ICD for secondary prevention in survivors of OHCA that, according to current
305
guidelines, depends on the exclusion of possible transient and reversible causes for VT/VF such as
306
acute myocardial ischemia or acute myocarditis.
M AN U
SC
RI PT
299
TE D
307
Funding Sources: the study was founded by the Target Projects RF-2013-02356762 grant of the
309
Italian Ministry of Health; the TRANSAC (translational arrhythmogenic cardiomyopathy) project,
310
2013, University of Padova, and the research grant BIRD (budget for integrated departmental
311
research) 2016, University of Padova, Italy.
AC C
EP
308
14
ACCEPTED MANUSCRIPT REFERENCES
313
[1] Priori SG, Blomstrom-Lundqvist C. 2015 European Society of Cardiology Guidelines for the
314
management of patients with ventricular arrhythmias and the prevention of sudden cardiac
315
death summarized by co-chairs. Eur Heart J.2015;36:2757-9.
RI PT
312
[2] Al-Khatib SM, Stevenson WG, Ackerman MJ, et al. 2017 AHA/ACC/HRS Guideline for
317
Management of Patients With Ventricular Arrhythmias and the Prevention of Sudden Cardiac
318
Death: Executive Summary: A Report of the American College of Cardiology/American
319
Heart Association Task Force on Clinical Practice Guidelines and the Heart Rhythm Society.
320
Heart Rhythm.2017;doi:10.1016/j.hrthm.2017.10.035.
M AN U
SC
316
321
[3] Anyfantakis ZA, Baron G, Aubry P, Himbert D, Feldman LJ, Juliard JM, Ricard-Hibon A,
322
Burnod A, Cokkinos DV, Steg PG. Acute coronary angiographic findings in survivors of out-
323
of-hospital cardiac arrest. Am Heart J.2009;157:312-8.
325
[4] Price S, Uddin S, Quinn T. Echocardiography in cardiac arrest. Curr Opin Crit Care.2010;16:211-5.
TE D
324
[5] Nolan JP, Neumar RW, Adrie C, et al. Post-cardiac arrest syndrome: epidemiology,
327
pathophysiology, treatment, and prognostication. A Scientific Statement from the
328
International Liaison Committee on Resuscitation; the American Heart Association
329
Emergency Cardiovascular Care Committee; the Council on Cardiovascular Surgery and
330
Anesthesia; the Council on Cardiopulmonary, Perioperative, and Critical Care; the Council on
331
Clinical Cardiology; the Council on Stroke. Resuscitation.2008;79:350-79.
333 334 335
AC C
332
EP
326
[6] Treibel TA, White SK, Moon JC. Myocardial Tissue Characterization: Histological and Pathophysiological Correlation. Curr Cardiovasc Imaging Rep.2014;7:9254. [7] Eitel I, Friedrich MG. T2-weighted cardiovascular magnetic resonance in acute cardiac disease. J Cardiovasc Magn Reson.2011;13:13.
15
ACCEPTED MANUSCRIPT 336
[8] Noc M, Fajadet J, Lassen JF, Kala P, MacCarthy P, Olivecrona GK, Windecker S, Spaulding C.
337
Invasive coronary treatment strategies for out-of-hospital cardiac arrest: a consensus
338
statement from the European association for percutaneous cardiovascular interventions
339
(EAPCI)/stent for life (SFL) groups. EuroIntervention.2014;10:31-7.
341
[9] Moss AJ, Schuger C, Beck CA, et al. Reduction in inappropriate therapy and mortality through
RI PT
340
ICD programming. N Engl J Med.2012;367:2275-83.
[10] Marcus FI, McKenna WJ, Sherrill D, et al. Diagnosis of arrhythmogenic right ventricular
343
cardiomyopathy/dysplasia: proposed modification of the task force criteria. Circulation
344
2010;121:1533-41.
SC
342
[11] Migliore F, Zorzi A, Perazzolo Marra M, Iliceto S, Corrado D. Myocardial edema as a
346
substrate of electrocardiographic abnormalities and life-threatening arrhythmias in reversible
347
ventricular dysfunction of takotsubo cardiomyopathy: Imaging evidence, presumed
348
mechanisms, and implications for therapy. Heart Rhythm.2015;12:1867-77.
M AN U
345
[12] Nucifora G, Muser D, Masci PG, et al. Prevalence and prognostic value of concealed structural
350
abnormalities in patients with apparently idiopathic ventricular arrhythmias of left versus right
351
ventricular
352
Electrophysiol.2014;7:456-62.
a
magnetic
resonance
imaging
study.
Circ
Arrhythm
EP
origin:
TE D
349
[13] Zorzi A, Perazzolo Marra M, Rigato I, et al. Nonischemic Left Ventricular Scar as a Substrate
354
of Life-Threatening Ventricular Arrhythmias and Sudden Cardiac Death in Competitive
355
Athletes. Circ Arrhythm Electrophysiol. 2016;9.pii:e004229.
AC C
353
356
[14] White JA, Fine NM, Gula L, Yee R, Skanes A, Klein G, Leong-Sit P, Warren H, Thompson T,
357
Drangova M, Krahn A. Utility of cardiovascular magnetic resonance in identifying substrate
358
for malignant ventricular arrhythmias. Circ Cardiovasc Imaging.2012;5:12-20.
359
[15] Baritussio A, Zorzi A, Ghosh Dastidar A, et al. Out of hospital cardiac arrest survivors with
360
inconclusive coronary angiogram: Impact of cardiovascular magnetic resonance on clinical
361
management and decision-making. Resuscitation.2017;116:91-97. 16
ACCEPTED MANUSCRIPT 362
[16] Marstrand P, Axelsson A, Thune JJ, Vejlstrup N, Bundgaard H, Theilade J. Cardiac magnetic
363
resonance imaging after ventricular tachyarrhythmias increases diagnostic precision and
364
reduces the need for family screening for inherited cardiac disease. Europace.2016;18:1860-5.
365
[17] Neilan TG, Farhad H, Mayrhofer T, et al. Late gadolinium enhancement among survivors of
368 369 370
RI PT
367
sudden cardiac arrest. JACC Cardiovasc Imaging.2015;8:414-23.
[18] Perazzolo Marra M, Lima JA, Iliceto S. MRI in acute myocardial infarction. Eur Heart J. 2011;32:284-93.
[19] Basso C, Perazzolo Marra M, Rizzo S, et al. Arrhythmic mitral valve prolapse and sudden cardiac death. Circulation 2016.132:556-66.
AC C
EP
TE D
M AN U
371
SC
366
17
ACCEPTED MANUSCRIPT 372
FIGURES LEGEND
373
Figure 1. Out-of-hospital cardiac arrest in a patient with obstructive coronary artery disease.
375
Out-of-hospital cardiac arrest occurring in a 64-year-old male with acute coronary syndrome.
376
Coronary angiography showing multiple vessels obstructive (≥70%) coronary artery stenosis
377
(arrows) affecting both the right (A) and the left (B) coronary artery. Cardiac magnetic resonance
378
showing transmural myocardial edema (three chambers, long axis view; T2-weighted sequences)
379
(C) and subendocardial late gadolinium enhancement (three chamber, long axis view; T1-weighted
380
inversion recovery post-contrast sequences) (D) affecting the infero-lateral left ventricular wall
381
which is tributary of the left circumflex coronary artery.
M AN U
SC
RI PT
374
382
Figure 2. Out-of-hospital cardiac arrest in a patient with obstructive coronary artery disease.
384
Out-of-hospital cardiac arrest occurring in a 57-year-old male with post-myocardial infarction scar.
385
Coronary angiography showing multiple vessels obstructive (≥70%) coronary artery stenosis
386
(arrows) affecting both the right (A) and the left (B) coronary artery. Cardiac magnetic resonance
387
showing no myocardial edema (three chambers, long axis view; T2-weighted sequences) (C) and
388
transmural late gadolinium enhancement (three chambers, long axis view; T1-weighted inversion
389
recovery post-contrast sequences) (D) affecting the entire left ventricular lateral wall which is
390
tributary of the left circumflex coronary artery.
AC C
EP
TE D
383
391 392 393
Figure 3. Out-of-hospital cardiac arrest in a patient with no obstructive coronary artery
394
disease.
395
Out-of-hospital cardiac arrest occurring in a 41-year-old male with biopsy-proven acute
396
myocarditis. Coronary angiography showing normal right (A) and left (B) coronary artery. Cardiac 18
ACCEPTED MANUSCRIPT 397
magnetic resonance showing myocardial edema (two chambers, long axis view; T2-weighted
398
sequences) (C) and transmural late gadolinium enhancement (two chambers, long axis view; T1-
399
weighted inversion recovery post-contrast sequences) (D) affecting the mid-apical inferior left
400
ventricular wall.
RI PT
401
Figure 4. Out-of-hospital cardiac arrest in a patient with no obstructive coronary artery
403
disease.
404
Out-of-hospital cardiac arrest occurring in a 29-year-old female with arrhythmic mitral valve
405
prolapse. Coronary angiography showing normal right (A) and left (B) coronary artery. Cardiac
406
magnetic resonance showing thickening and prolapse of both mitral valve leaflets (C,D), no
407
myocardial edema (three chambers, long axis view; T2-weighted sequences) (C) and evidence of
408
mid-myocardial late gadolinium enhancement (three chambers, long axis view; T1-weighted
409
inversion recovery post-contrast sequences) (D) affecting the basal infero-lateral left ventricular
410
wall behind the prolapsing posterior mitral valve leaflet.
411
TE D
M AN U
SC
402
Figure 5. Out-of-hospital cardiac arrest in a patient with no obstructive coronary artery
413
disease.
414
Out-of-hospital cardiac arrest occurring in a 36-year-old male with an isolated non-ischemic left
415
ventricular scar. Coronary angiography showing normal right (A) and left (B) coronary artery.
416
Cardiac magnetic resonance showing no myocardial edema (four chambers, long axis view; T2-
417
weighted sequences) (C) and the presence of subepicardial late gadolinium enhancement (four
418
chambers, long axis view; T1-weighted inversion recovery post-contrast sequences) (D) affecting
419
almost the entire left ventricular lateral wall.
AC C
EP
412
420 421
19
ACCEPTED MANUSCRIPT TABLES
422
Table 1: clinical characteristics of the study population Coronary stenosis ≥ 70% N=18 17 (94%) 57 (54-71)
NO Coronary stenosis ≥ 70% N=26 19 (73%) 44 (27-59)
P
0.11 Gender (male) <0.001 Age (years) Risk factors Hypertension 13 (72%) 5 (19%) 0.001 Diabetes 6 (33%) 2 (8%) 0.05 Smoking 8 (45%) 4 (15%) 0.04 Dyslipidemia 8 (44%) 5 (19%) 0.09 Family history for CAD 4 (22%) 4 (15%) 0.70 Personal history Ischemic heart disease 4 (22%) 0 0.03 Non-ischemic heart disease 0 3 (12%) 0.25 Chest pain within 1 month 6 (33%) 1 (4%) 0.02 Syncope/pre-syncope 0 3 (12%) 0.25 Characteristics of cardiac arrest Cardiac arrest in public place 11 (61%) 9 (35%) 0.06 Witnessed cardiac arrest 17 (95%) 23 (88%) 0.63 Bystander CPR 11 (61%) 22 (85%) 0.09 ST-segment elevation post 7 (39%) 0 <0.001 resuscitation Echocardiography at admission 42 (33-50) 50 (47-56) 0.014 LV EF (%) 69 (60-84) 0.40 71 (56-96) LV EDVi (ml/m2) <0.001 LV regional WMA 17 (94%) 8 (31%) 2 (11%) 5 (19%) 0.68 RV dilation 2 (8%) 0.68 RV dysfunction 2 (11%) RV regional WMA 1 (6%) 3 (12%) 1.0 Mitral valve regurgitation ≥ 2 (11%) 3 (12%) 1.0 moderate 5 (19%) 0.06 Mitral valve prolapse 0 Aortic valve stenosis ≥ moderate 0 0 5.1 (1.5-82) 1.2 (0.29-9.9) 0.02 Troponin I peak (umol/L) Coronary angiography At admission 12 (67%) 8 (31%) 0.41 Time from admission 1 (0-3) 2 (1-3) 0.55 Single vessel disease 5 (28%) 0 Multiple vessel disease with 5 (28%) 0 culprit lesion Clinical diagnosis (no CMR) Structurally normal heart* 0 12 (46%) <0.001 Ischemic heart disease 18 (100%) 1 (4%) Acute myocarditis 0 3 (12%) Cardiomyopathy# 0 5 (19%) Mitral valve prolapse 0 5 (19%) * including Brugada syndrome, idiopathic catecholaminergic ventricular tachycardia, long QT syndrome, and short QT syndrome # including non-ischemic dilated cardiomyopathy, arrhythmogenic cardiomyopathy and hypertrophic cardiomyopathy
AC C
EP
TE D
M AN U
SC
RI PT
423
424 425 426 427 428 429 430
CMR=cardiac magnetic resonance; ICD=implantable cardioverter defibrillator; LV=left ventricular; RV=right ventricular; TAPSE=tricuspid annular plane systolic excursion; WMA= wall motion abnormalities; TTS=Tako-tsubo syndrome
20
ACCEPTED MANUSCRIPT Table 2: cardiac magnetic resonance findings in the study population P
3 (17%) 3 (17%) 7 (39%) 5 (28%)
18 (69%) 6 (24%) 2 (8%) 0
<0.001
7 (39%) 6 (33%) 2 (11%) 3 (17%) 17 (94%) 1 (6%) 0 2 (11%)
20 (77%) 3 (12%) 2 (8%) 1 (4%) 5 (19%) 4 (15%) 1 (4%) 3 (12%)
6 (33%) 0
20 (77%) 4 (15%)
1 (6%) 11 (61%)
1 (4%) 1 (4%)
0 0 7 (39%) 11 (61%) 1 (6%)
7 (27%) 17 (65%) 1 (4%) 1 (4%) 2 (8%)
<0.001
7 (27%) 2 (8%) 4 (15%) 7 (27%) 3 (12%) 1 (4%) 2 (8%)
<0.001
RI PT
NO Coronary stenosis ≥ 70% N=26
0.03
<0.001 0.52 1.0 1.0
SC
M AN U
0 18 (100%) 0 0 0 0 0
EP
Morpho-functional parameters LV systolic function Normal Mild impairment Moderate impairment Severe impairment LVEDV Normal Mild dilation Moderate dilation Severe dilation LV regional WMA RV systolic impairment RV dilatation RV regional WMA LV Myocardial edema Absent Patchy, midmyocardial or subepicardial Subendocardial Transmural LV LGE Absent Midmyocardial or subepicardial Subendocardial Transmural RV LGE CMR diagnosis Structurally normal heart* Ischemic heart disease Acute myocarditis Cardiomyopathy# Mitral valve prolapse with LGE Tako-Tsubo syndrome Non-ischemic LV scar
Coronary stenosis ≥ 70% N=18
TE D
431
0.002
1.0
CMR=cardiac magnetic resonance; LGE=late-gadolinium enhancement; LVEDV= left ventricular end-diastolic volume; LV=left ventricular; RV=right ventricular; WMA= wall motion abnormalities;
436
# including non-ischemic dilated cardiomyopathy, arrhythmogenic cardiomyopathy and hypertrophic cardiomyopathy
437
AC C
432 433 434 435
21
ACCEPTED MANUSCRIPT 438
Table 3 – comparison between clinical and cardiac magnetic resonance diagnosis Suspected diagnosis (pre-CMR) Structurally normal heart (N=12)
Acute myocarditis (N=3)
Cardiomyopathy (N=5)
SC
Mitral valve prolapse (N=5)
M AN U
Agreement
18/19 (95%) 1/3 (33%)
4/5 (80%)
3/5 (60%)
18/18 (100%) 15/26 (58%) 33/44 (75%)
EP
TE D
CMR=cardiac magnetic resonance; LGE=late gadolinium enhancement; LV=left ventricular
AC C
439
Agreement 7/12 (58%)
RI PT
Ischemic heart disease (N=19)
Final diagnosis (after-CMR) Structurally normal heart = 7 Acute myocarditis = 2 Cardiomyopathy = 1 Non-ischemic LV scar = 2 Ischemic heart disease = 18 Tako-tsubo syndrome = 1 Acute myocarditis = 1 Ischemic heart disease = 1 Cardiomyopathy = 1 Cardiomyopathy = 4 Ischemic heart disease = 1 Mitral valve prolapse with LGE = 3 Cardiomyopathy = 1 Acute myocarditis = 1 Coronary stenosis ≥70% group NO coronary stenosis ≥70% group Overall agreement
22
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT